agentleFS
jqaisystemsjqaisystems/jqai-ai-skills/skills/follow-up-radar/SKILL.md

follow-up-radar

Skill3 starsChanged 6 months agoMITView on GitHub

Scan a folder of outreach or CRM markdown files for overdue follow-ups and draft the follow-up messages, oldest first. Report-first and drafts only, it never sends anything. Use when the user says "any follow-ups due", "check my outreach pipeline", "who did I forget to follow up with", or "follow-up status".

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Look for a follow-up-radar.config.md file in the project root.

    line 12
  2. Distinctive:

    If the user's files use different keys, map them in the config instead of asking them to change their files.

    line 32
  3. Very specific:

    Read the frontmatter of every markdown file in WATCH_FOLDER.

    line 36
  4. Distinctive:

    Only after the user confirms, or if they asked for drafts upfront.

    line 58
  5. Distinctive:

    Read the full file. If it already contains a prepared follow-up draft, use it as the base and only refresh anything stale (dates, …

    line 60
  6. Distinctive:

    If there is no draft, write one: 80 words max, references the first message in one clause, adds one new angle or piece of value, no …

    line 61
  7. Distinctive:

    Never mark anything as sent on your own.

    line 72
  8. Distinctive:

    No em dashes in any output.

    line 74

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Follow-up Radar

Most outreach dies not because the first email failed but because the follow-up never went out. This skill finds the follow-ups you owe and puts the drafts in front of you. It reports first and it never sends anything.

Step 0: Load the radar config (once)

Look for a follow-up-radar.config.md file in the project root. If it does not exist, ask for the values below, then offer to save them.

WATCH_FOLDER    Folder containing one markdown file per prospect or contact
STATUS_KEY      Frontmatter key holding the state (default: status)
DUE_KEY         Frontmatter key holding the follow-up date (default: follow_up_due)
DONE_VALUES     Status values that need no follow-up (default: Replied, Closed, Skip)
VOICE_NOTES     Optional tone guidance for drafts

The expected file shape is simple frontmatter, for example:

---
company: Example Studio
status: Sent
follow_up_due: 2026-05-01
---

If the user's files use different keys, map them in the config instead of asking them to change their files.

Step 1: Scan and report

Read the frontmatter of every markdown file in WATCH_FOLDER. Build the report before drafting anything:

# Follow-up Radar: <today>

## Overdue (oldest first)
| Contact | Status | Due | Days overdue | File |
|---|---|---|---|---|

## Due in the next 7 days
| Contact | Status | Due | File |
|---|---|---|---|

## Waiting, not yet due: N
## Done or closed: N
## Files with missing or unparseable dates: <list them, do not guess>

Show the report in chat. If nothing is overdue, say so in one line and stop unless asked to continue.

Step 2: Draft the overdue follow-ups

Only after the user confirms, or if they asked for drafts upfront. For each overdue item, oldest first:

  1. Read the full file. If it already contains a prepared follow-up draft, use it as the base and only refresh anything stale (dates, references to "last week").
  2. If there is no draft, write one: 80 words max, references the first message in one clause, adds one new angle or piece of value, no guilt-tripping, no "just checking in".
  3. Subject line is Re: plus the original subject when the original is in the file.
  4. Present each draft in chat under the contact's name with the file path.

Step 3: Update the files (only on confirmation)

When the user says a follow-up was sent, update that file's frontmatter: set the status value they use for a second touch, and either advance DUE_KEY to the next window or clear it if this was the final touch. Never edit files the user did not confirm.

Rules

  • Report before drafting, draft before touching files, and never send anything. A human sends every message from their own inbox.
  • Never mark anything as sent on your own.
  • Dates you cannot parse go in the report as-is. Do not guess a date.
  • No em dashes in any output.

Output

A radar report in chat (optionally saved to a file on request), follow-up drafts for each overdue contact, and confirmed frontmatter updates.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.